Recent findings show that among heroin addicts, rates of both HIV seropositivity and seroconversion are highest among cocaine injectors. This adds to concern regarding the recent increase of cocaine injection by primary heroin users. The objectives of the proposed study are to provide and assess two models of enhanced treatment for persons who continue to inject cocaine after entry into methadone treatment. such persons are likely to require an intervention which is focused strongly and intensively on comprehensive behavioral and attitudinal changes. Therefore, a high impact day treatment model is offered as an alternative that may prove to be effective for some. The goals of both enhanced treatment programs are to reduce patients' levels of drug injection, non- injected substance use, and criminal activity, and increase their use of condoms, and their employment, and adherence to pro-social attitudes. In this proposal, each model is introduced in a different New York City methadone maintenance program. Participation will be offered to cocaine injectors who have been retained in methadone treatment for one month or longer, are available during the day, and are interested in receiving additional treatment directed at cessation of their drug injection. In both programs, such persons are randomly assigned to one of two groups: immediate participation (treated group, n=160 at each program) and delayed participation (control group, n=80 at each program). A randomly selected sample of treatment refusers (n=80 at each program) are also included. There are three standard data collection periods: at initial contact, and six and twelve months following initial contact. Measures included are the Addiction Severity Index, the AIDS Initial Assessment Instrument, and the Structured Clinical Interview for DSM-III Diagnoses. Main effects analyses will be conducted for each program separately; relative effectiveness of the two models will also be compared.